Data

Tools

Indexes

Applications

Experiments

Open Source Science

https://github.com/deepskies/deepmhd

https://github.com/deepskies/deepmhd

Science Score: 23.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ○
    codemeta.json file
  • ○
    .zenodo.json file
  • ✓
    DOI references
    Found 4 DOI reference(s) in README
  • ✓
    Academic publication links
    Links to: arxiv.org
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (5.3%) to scientific vocabulary
Last synced: 5 months ago · JSON representation

Repository

Basic Info
  • Host: GitHub
  • Owner: deepskies
  • License: bsd-3-clause
  • Language: Jupyter Notebook
  • Default Branch: master
  • Size: 705 KB
Statistics
  • Stars: 2
  • Watchers: 7
  • Forks: 3
  • Open Issues: 0
  • Releases: 0
Created almost 7 years ago · Last pushed over 6 years ago

https://github.com/deepskies/deepmhd/blob/master/ 

### deepmhd

These notebooks are provided to reproduce all data-containing figures and all results from 

```Do Androids Dream of Magnetic Fields? Using Neural Networks to Interpret the Turbulent Interstellar Medium```

by J. E. G. Peek and Blakesley Burkhart, accepted ApJL, https://arxiv.org/abs/1905.00918

With these notebooks, and the associated data [here](https://doi.org/10.7910/DVN/UKOPYP), you should be able to 

- extract normal and Fixed Fourier Power (FFP) images from the turbulent simulations and save them as files for training and test data
- train networks to classify these images
- evaluate the networks (either your own trained networks, or the ones provided by us at [here](https://doi.org/10.7910/DVN/UKOPYP))
- run the saliency map analysis
- make figures 1 and 3 (without annotation)

### Software Requirements

These notebooks have a few dependencies:

- Python 3
- numpy 1.14.5
- tensorflow 1.6
- keras 2.2.4
- sci-kit-image 0.13
- matplotlib 2.1.2
- tqdm

### Hardware Requirements

There should not be strong hardware requirements to make figures and evaluate networks. To train the networks we used a NVIDIA Tesla P100 GPU, and we do suggest GPU acceleration of some kind for timely network training.

Owner

  • Name: Deep Skies Lab
  • Login: deepskies
  • Kind: organization
  • Email: deepskieslab@gmail.com

Building community and making discoveries since 2017

GitHub Events

Total
  • Watch event: 1
  • Fork event: 1
Last Year
  • Watch event: 1
  • Fork event: 1